M: I don't want you to do this. So don't start any prep work or anything like that. This is just a conceptual idea. Would you be capable of aggregating all of the publicly available material on any given representative, and then filtering it in a way that would give you, or rather give the user, a psychological breakdown based on their public persona that is not necessarily a complete personality diagnosis, but instead just an analysis of the person they present compared to their actions. Without trying to make anything absolute, the concept this stems from would be less of an absolute measurement and more of a comparative diagnosis of the almost schizophrenic tendency to campaign and broadcast one thing while voting or acting another way, while out of the public eye. I mean, I'm getting seriously worried about some of the members of Congress in terms of their ability to understand that they are representing more than themselves in any given chamber.
G: [Non-text content: thoughts; content_type=thoughts]
G: [Non-text content: reasoning_recap; content_type=reasoning_recap]
G: Yes, with limits.
I could help build a public-record “persona vs. conduct” analysis for a representative using:
public speeches, campaign pages, press releases, interviews, social media, floor statements, committee behavior, roll-call votes, bill sponsorships, donor/industry ties, ethics records, and local impacts.
But I would not frame it as a psychological diagnosis or “schizophrenic tendency.” That veers into mental-health labeling from public behavior, which is unreliable and unfair. A safer, stronger frame would be:
Public Persona vs. Governing Behavior Audit
It could measure things like:
Rhetoric/action divergence: what they say they value versus how they vote or govern.
Constituency alignment: whether their actions match district needs, demographics, and stated promises.
Role comprehension: whether they behave like a representative of a plural constituency or like a factional/media actor.
Consistency over time: whether shifts are explained, opportunistic, or hidden.
Public/private asymmetry: whether they use broad moral language publicly while taking narrower or contradictory institutional actions.
The output would be more like a structured political-behavior profile than a psych profile:
“This member presents as X, but their voting/sponsorship/committee behavior more often reflects Y. The gap is strongest on A and weakest on B. The available record supports calling this strategic inconsistency / factional representation / performative populism, but not a clinical judgment.”
So: yes, capable in principle — but the useful version is not “diagnose the person.” It is audit the contradiction between representation-as-claimed and representation-as-performed.
M: Yeah, diagnosis was perhaps the wrong choice of words. This would be more of a comparative index kind of situation. I'm having trouble with the language because this is more of a feeling at the moment than something I've actually talked out. The concern is like an ongoing one that seems to have amplified over the last ten years or so. Like, before then, the sensation was that people would talk a big game but would quietly go against their stated national positions in order to maintain positive outcomes for their voting constituency. Like, I think Mitch McConnell would probably be a good example of that, at least in the past, where he had philosophical statements which were underpinned with a concern for Kentucky getting what he felt it deserved, whether it did or not. I think it was Kentucky. But the point is that there seems to be more of a lack of concern about either national or local societal maintenance with a turn towards almost an effort towards endgame victory over the people who are trying to keep the country going or something like that. Am I making sense? It's this feeling that the... I'm gonna use the term hypocrisy as a non-judgmental form here because that's not my intent. I'm talking about when the party line contradicts actual voting record. There seems to be an abandonment of the idea of doing that for the local constituency, and instead almost a desire to actually undercut government itself in loyalty and betrayals. Again, I'm using that terminology because it's the easiest, not because I'm trying to assign any kind of moral value there. Saying you'll do one thing and doing another politically is sometimes the only way to get things done in the past. But now it's like the things that are held to are not the things that would actually make things better for any citizens, but instead push forward an agenda based on the idea that somehow, you know, if you get everything you want philosophically, then afterwards magically the citizenry will be ideally taken care of in a way that isn't actually expressed. I'm sorry if this has been kind of a ramble. Like I said, I'm trying to express an idea that's largely just been a growing feeling.
G: You are making sense, and I think the difficulty you're having is that you're trying to describe a change in the unit of political loyalty, not merely an increase in hypocrisy.
Traditionally, political scientists have talked about tensions like:
What you're circling around sounds like a possible fourth category:
Your impression is that this fourth category has become weaker relative to the others.
That's different from saying politicians have become "more hypocritical." Hypocrisy is about inconsistency between words and actions. What you're describing is a shift in what appears to constrain their actions.
Earlier, your perception is that a politician might:
The modern pattern you're describing feels more like:
Whether that perception is correct in any specific case is something that would have to be evaluated with evidence. But as a conceptual distinction, it's coherent.
I also think you're trying to distinguish between two kinds of inconsistency:
Instrumental inconsistency
Structural inconsistency
Those are analytically different things.
If I were designing the kind of framework you mentioned earlier, I probably wouldn't score "hypocrisy" at all. I'd score independent dimensions, for example:
| Dimension | Question |
|---|---|
| Promise consistency | Do public commitments match later actions? |
| Constituency alignment | Do actions plausibly benefit the represented district or state? |
| National governance alignment | Do actions appear aimed at improving national governance, even when politically costly? |
| Institutional stewardship | Does the official act in ways that preserve the effectiveness and legitimacy of governing institutions? |
| Partisan prioritization | When these goals conflict, which one tends to prevail? |
That would let you compare representatives without making claims about their psychology.
One thing I'd add, because it would make the framework stronger, is to avoid assuming that institutional conflict is inherently evidence of institutional disregard. Sometimes legislators genuinely believe that weakening or restructuring part of government is itself the best way to improve governance. The analytical question then becomes: is that belief reflected consistently in their actions, and are they transparent about the tradeoffs? A framework that distinguishes "this person is trying to reduce government because they think it improves government" from "this person says one thing and appears to pursue a different objective" would be more robust and less dependent on the observer's political perspective.
So I don't think you're really describing a feeling anymore. I think you're groping toward a model of representational priorities—trying to identify what level of obligation a public official seems to treat as paramount when their commitments come into conflict. That's a concept that can be operationalized without requiring any assumptions about the person's internal motives.
M: Yeah, it's like, back in the day, the justification for being against any kind of social safety net was a kind of twisted idea of being for your own good. Like, you could understand what was being said even if you completely disagreed with it. The idea that support made people weaker and reluctant and cost people money. Now, studies have shown that was wrong, but people wanted to believe it was that easy and straightforward, and so it made sense. Now it's like, the idea of being for your own good isn't even that kind of simplistic statement, but instead just is people saying, put me in charge because you don't like them, and we should end welfare. It's for your own good, don't worry about why, just trust me. Very much a 80s, I think Glengarry Glen Ross, I really need to watch that. But the idea of greed is good, if I'm in charge, things will be better, so you should believe me because I speak well, kind of attitude. Not because there's any actual philosophy behind it, except that there's always tomorrow to fix things, and right now I want to be in charge, and make sure that I stay in charge. And I'm not talking about the system itself. I'm talking about the people who are trying to undercut the system. Like the ones who want to stay in charge because they are in charge, and they want to change the rules so that they stay in charge. Like the gerrymandering shit that's been going on. It's not because they've been doing a better job, but because they know that they haven't been doing a better job. If they were doing a good job, people would appreciate that instead of having to change the rules so that they can maintain the ability to change the rules. It's a degree of just self-interest. Or even more disturbingly, like, I don't know, what's the quick term for it? Not self-interest, but cronyism. Like, everyone defining their own self-interest as being in this one guy, or one set of people, and not sacrificing anything for their constituents or even themselves, but this person under this unconfirmed belief that somehow everything will be better as long as the cult leader is in charge. And that the priority is to make sure that they stay in charge. Or just this idea even, not even a person, but just the idea that victory is the thing.
G: I think you've sharpened the distinction considerably. What you're describing is less about what policies someone supports and more about what appears to be the ultimate objective that organizes those policies.
To put it abstractly, you could imagine several different "organizing principles" behind a politician's behavior:
Every politician has some mixture of these. The question is which one seems to dominate when they come into conflict.
Where I would be cautious is in moving from observations like gerrymandering or institutional rule changes to concluding that the motivation must be "they know they haven't been doing a better job." There are other possibilities. Some politicians sincerely believe demographic or institutional trends structurally disadvantage their coalition and therefore justify aggressive procedural tactics. Others may view politics as an existential contest in which losing power would permanently prevent the policies they see as vital. Those beliefs don't make the tactics necessarily wise or justified, but they do mean we shouldn't assume a single internal motivation.
That said, your broader concern has been discussed by political scientists under ideas like constitutional hardball, negative partisanship, and democratic backsliding. The common thread is that political competition shifts from "winning within agreed rules" toward "changing or exploiting the rules because victory itself becomes paramount."
One thing I find interesting in your formulation is that you're not saying, "I dislike politicians who disagree with me." You're proposing a hierarchy:
That's actually a more analytically useful framework than starting with ideology.
For example, in principle, you could imagine two legislators who both vote against expanding welfare:
Those are different patterns, even if the roll-call vote is identical.
Your idea also suggests something measurable. Rather than asking, "Is this representative good?" you could ask:
Over dozens or hundreds of decisions, that starts to look less like a judgment of a single vote and more like a behavioral profile.
One thing I'd preserve in such a framework is symmetry. If it's designed so that it can reveal this pattern in officials of any party, then it's much more likely to capture the phenomenon you're interested in rather than simply reflecting disagreement with particular policies. The value of the framework would come from asking the same questions—and applying the same standards—regardless of whether the representative is someone you generally support or oppose. That way, if someone consistently sacrifices short-term political advantage to uphold their stated principles or institutional commitments, the framework can recognize that too.
M: Okay, two different responses, both of which are saying that, which are responses to your talking about how you can not know motive. I'm looking at it from at least two different directions where I disagree. One, these are public officials, they have people giving them this information for governing reasons, if not strictly political reasons. There are polls, there are methods, etc., etc. So, often, if their motive is not something, then that would imply that they are completely out of touch despite that, and have already kind of shown that they don't know what the fuck they're doing. Which kind of puts them into the power for power's sake category by default. And then you're falling again into your usual idea of not being able to know what lies in the hearts of man kind of thing. You're taking that to such an extreme that it derails the conversation, where the understanding is generally that, yes, everything is a variable up in the air, everything can change. But kind of in the same way that... I was gonna make a metaphor, but there's a difference between the absolute certainty that you seem to be functioning under and my 99.9% accuracy, while allowing that I could be wrong. I always allow that I could be wrong. So you don't need to quibble about that. And that's kind of what we're doing anyway, is this probability thing. So the idea that, hey, we have to worry about probabilities when we're talking about probability, it's a little bit redundant. Talking about the likelihood, that kind of thing. If you're trying to win power for, or money for Kentucky, then you don't vote for things that keep money from going to Kentucky. I mean, that right there indicates that something else is going on. And yeah, in this situation, whether I agree with the policies is beside the point. I'm not even asking as a final question if I agree with the policies. I mean, okay, Marjorie Taylor Greene is not someone that I would be concerned about right now in terms of the measurements we're talking about, because even though I think that she has batshit insane stances on certain things, they are consistently batshit insane. The fact that Republicans turned on her is kind of indicative of where my concern is. The votes are not... their attitudes and votes are not consistent even with themselves lately, and I think could be shown to be shaped around particular attractors that have nothing to do with either their personal local constituency or an ideologically consistent item. I'm just looking to describe that through some form of analysis. Probably a good way to do this would be to run a test run on someone who's already retired and is well documented.
G: I think this gets at an important distinction between epistemic caution and analytical paralysis, and I agree that I was leaning too far toward the latter for the discussion you were trying to have.
You're not asking me to prove a person's internal mental state. You're asking whether, from a sufficiently large body of public behavior, we can infer the most likely organizing principle behind that behavior, while explicitly accepting that it's an inference rather than certainty.
That's a perfectly ordinary form of analysis. We do it all the time in other domains.
For example:
None of those claim omniscience. They claim that repeated patterns become informative.
I also think your Kentucky example illustrates the kind of thing you'd want to measure. Suppose a representative repeatedly says:
"My first responsibility is my constituents."
Then suppose a series of votes consistently produce outcomes that:
After enough observations, it's reasonable to say something like:
"The observable evidence suggests that constituency benefit is not the dominant predictor of this representative's behavior."
That's not a statement about their soul. It's a statement about predictive power.
I actually think the statistical framing may help here.
Imagine assigning competing hypotheses:
Then you examine hundreds of observations and ask:
Which hypothesis best predicts future actions?
Now you're doing something much closer to model fitting than psychology.
That strikes me as considerably more rigorous.
I also liked your example of entity["politician","Marjorie Taylor Greene","U.S. Representative from Georgia"] because it highlights something subtle.
You're not saying consistency is good.
You're saying consistency is diagnostically useful.
Someone can consistently advocate positions you strongly oppose, yet still exhibit a stable decision-making framework.
Your concern is almost the opposite:
When someone's public rationale changes from issue to issue, but one hidden variable seems to predict everything.
That's a much more interesting question.
If we were designing this from scratch, I'd avoid trying to classify people morally and instead try to discover latent attractors.
That's actually the phrase that jumped out at me from what you said.
Instead of asking:
"Is this representative hypocritical?"
Ask:
"What variable best explains this representative's decisions?"
Possible attractors might include:
The analysis wouldn't begin with any assumption about which one dominates. It would ask which one has the highest explanatory power.
I also agree with your last point about choosing a retired figure.
That's methodologically cleaner because:
Someone like entity["politician","John McCain","U.S. Senator from Arizona"], entity["politician","Robert Byrd","U.S. Senator from West Virginia"], entity["politician","Tip O'Neill","Speaker of the U.S. House"], or entity["politician","Paul Ryan","Former Speaker of the U.S. House"] could make useful test cases because each has an extensive public record with recognizable patterns and enough distance that the exercise is less likely to become an argument about today's headlines.
What interests me most about your proposal is that it isn't really trying to answer "Who is right?" It's trying to answer, "What consistently drives this person's behavior?" If you can build a framework that reliably distinguishes "constituency-driven," "ideology-driven," "institution-driven," and "power-driven" patterns across people with very different politics, then you've got something that's useful beyond any single politician or party. It becomes a way of characterizing governing behavior rather than adjudicating political beliefs.